Generative AI for hydrodynamical simulations:

2D, 3D, or 6D galaxy models?

SKA research at
Zurich University of Applied Sciences (ZHAW)

Centre for Artificial Intelligence (CAI)
Institute for Business Information Technology (IWI)
June 10, 2024
contact_qr.png Philipp Denzel, Frank-Peter Schilling, Elena Gavagnin

Slides on my website

https://phdenzel.github.io/

Outlook

Recap:
Generative models
for map-to-map translation

Dataset from IllustrisTNG

  • projected IllustrisTNG galaxies
  • 6 domains:
    • dark-matter, stars, gas,
      HI, temperature, magnetic field
  • ∼ 2'000 galaxies, (across 6 snapshots)
  • ∼ 360'000 images
  • each galaxy \(\ge\) 10'000 particles
  • augmented: up to 5x randomly rotated
  • scale: 2 dark-matter half-mass radii

Dataset from IllustrisTNG

  • projected IllustrisTNG galaxies
  • 6 domains:
    • dark-matter, stars, gas,
      HI, temperature, magnetic field
  • ∼ 2'000 galaxies, (across 6 snapshots)
  • ∼ 360'000 images
  • each galaxy \(\ge\) 10'000 particles
  • augmented: up to 5x randomly rotated
  • scale: 2 dark-matter half-mass radii

Generative model architectures


Benchmark of generative models we're investigating and comparing:

cGANs

Figure 1: pix2pix scheme

Figure 2: cGAN(Gas) → DM: data, prediction, and ground truth (from top to bottom)

Score-based diffusion (SDM)

Figure 3: Score-based diffusion: Song et al. (2021)

Noise schedule

Inversion by Direct Iteration (InDI)

Figure 4: InDI's iteration scheme: Delbracio & Milanfar (2023)

Diffusion Mamba (DiM)

Figure 5: DiM architecture Teng et al. (2024)

From 2D to 3D models

  • observations inherently have 2D spatial resolution
  • astrophysical structures are inherently 3D
  • modelling difficulties:
    • inherent 3D features, different 2D perspectives
    • degeneracies

Inherent 3D shapes

Degeneracies


original image




reconstruction


Point-cloud models

Point-cloud experiments

"Phase-space-point" models

References

Created by phdenzel.